Machine learning for the characterisation and design of battery electrodes

Thur, 18 April 2024, 11.00-12.00

Battery companies want to know the relationship between their manufacturing parameters and the performance of the resulting cells so that they can optimise their products for particular applications, reduce costs, and improve yield. The literature contains many examples of physics-based models of the various manufacturing processes (including mixing, coating, drying, and calendaring), but these systems are hugely complex, and as a result they are expensive to simulate and hard to validate.

Recent advances in generative machine learning (ML) methods have allowed the relationship from manufacturing parameters to microstructure to be directly learned from data.

In this talk I will present a modular approach to the cell optimisation cycle that makes use of these ML methods, in combination with GPU accelerated metric extraction (TauFactor 2), electrochemical cell simulation (PyBaMM), and Bayesian optimisation. In addition, I will be introducing a new kintsugi SEM imaging method for accurately observing the nanostructure of the carbon binder domain; “VoxCel” an open-source, voxel-based, GPU-accelerated, multi-physics cell simulation; ML methods for generating 3D data from 2D images, as well as, inpainting artefacts in image data; and a data fusion method for combining multi-modal datasets using GANs. Lastly, I’ll present a webapp that normalises the data obtained from testing cells in a lab for easy comparison to commercial cells: cell-normaliser

We are always looking for new collaborations and new data so please get in touch! If you’d like to use any of our suite of open-source tools, then head to our website: https://tldr-group.github.io

We’ve also just spun-out a company from Imperial, called Polaron, to bring these tools to market. Check out our website (www.polaron.ai) and get in touch: info@polaron.ai